First, I want to thank the authors for addressing my comments and suggestions and including Appendix A. The authors are persistent in arguments that attempt to justify their claim of accurately assessing albedo for cases where terrain and sensor tilts are unknown, and often persistence leads to progress. What is most impressive about this team is their ambition to tackle this widely known and challenging problem. Field studies such as this provide valuable information that should be published. This study is very much in its infancy and deserves opportunities for improved field methods as revealed by peer reviews. Unfortunately, the current field techniques do not have broad application, largely by their limitations to necessary insolation conditions and lack of in situ, concurrent measurements that identify the solar zenith impact on surface albedo. For example, the authors are aware of sensors with narrower field of view applied in studies of spectral and zenith angle dependence of albedo, but did not attempt to deploy concurrently during the field season. The authors should collect more evidence to strengthen their arguments and resubmit a new manuscript.
With regards to grammar, the authors often place two-or-three-sentence “paragraphs” that could easily be merged with larger paragraphs. A paragraph should have at least five sentences in my opinion.
Regarding improvements from the first submission, the Introduction is much better, the methods section still focusses too much on well-established equations and laws (which can probably be referenced instead of explicitly included). For example section 2.4.1 might be combined into 2.4.2 by simply stating the references and the Fhor = .. equation. There are still a few assumptions made that might be better explained or referenced, but much better than the first submission. Section 2.5 can be reduced given the nicely designed flow chart.
Regarding figures, most authors recognize that figure in color will copy (in gray scale) poorly if certain colors are used and no markers or dashed-dotted-line thickness is applied, so all graphs are subject to this limitation. On the other hand, the audience today has access to read journals online on computers and mobile devices that detect the color. B&W figures are improved, and the addition of Fig. 7 is helpful.
With regards to justification of techniques, their claims of challenges in measuring tilt of sensors with varying azimuth are warranted, but definitely not impossible to overcome with modern tilt sensors. Just pull out your smart phone and you will find this technology using the compass. That said, very few, if any vendors provide such sensors that integrate nicely with reliable AWS units on the market.
Introduction is much improved although scratching my head at some sentences, such as the second sentence. There is still some “fluff” that can be eliminated such as language “more or less” or “as well as.”
The discussion still needs a bit of work. The authors clearly state the limitation of requiring 2 hours of clear sky daylight conditions (presumably also this should be a SZA < 50). It would be nice to see a global map of % days with 2 hours or more clear skies to better inform readers of potential for application. There are many areas of the world that have distinct dry seasons with high altitude glaciers that would qualify. The authors do a good job summarizing the limitation of SZA < 50. However, the authors should focus on the potentially exciting applications; this technique could be adapted for tropical glaciers and seasonal snow-fields and this should be an area of future work. Also, it is clear that results indicate the need for corrections on slopes more orthogonal to SZA (aka the southwest facing direction receives more direct irradiance), and that albedo should be less than that measured by over 10 percent, which is very significant for mass balance studies. Perhaps this is already understood in the literature, but the authors offer a unique method that needs further testing and validation in other locations. Validation could include setting up stakes or down-facing IR thermometry (or thermal imaging) to record depth and surface temperature distribution compared to those of energy-mass balance models incorporating the authors’ approach to albedo correct.
The conclusion could also include a more positive note of potential application and future testing of the techniques for snow/ice in lower latitudes, especially tropical glaciers where a great deal of work is ongoing.
p.7 l.16 Already defined AWS.
p.7 l.18 at what height above the snow surface? “measuring also air temperature, wind speed and direction, relative humidity and air pressure to determine mass- and energy balance of the glaciers”
p.8 l.23 better wording, and using the word calculated too much. Try this “In the present case the model <estimated solar irradiance> for the location…”
p.13 l.5 constant azimuth angle of what? the pyranometer (please state this).
p.13 l.10 This assumption is probably fine, although we know that at high thetas (SZA) there is some specular reflection leading to some direct component, but it is negligible perhaps since the total irradiance is so low and insignificant for energy budget studies.
p.15 l.14 Strangely worded sentence and difficult to understand. How about linking to the previous sentence “…for a day, and our analysis corrects albedo for periods where SZA < 50. If this is the case, then this needs to be stated as a limitation of the results.
p.15 l.19 to p.17: Section 2.5, In my opinion, all you need is your nicely compiled flow diagram and an explanation of Step C (especially the factor C). There is no need for subsection headings here, eliminate all but explain Step C.
p.20 l.4 “In Figure 2 as well as in Figure 9 ” Try to avoid the phrase “as well as”, simply state “and”
p.20 l.15 maintained at field works? You mean maintained during field visits? How often did you visit the sites?
Appendix A: Why thetat AND thetatilt, why not just combine into thetat throughout since they are defined exactly the same?